Scooter frame welding robot control method and welding robot

Abstract

The application discloses a scooter frame welding mechanical arm control method and a welding mechanical arm and relates to the technical field of welding robot control, which comprises the following steps: identifying a curvature mutation area based on the curvature change rate of a welding seam track, performing dynamic feedforward speed planning, obtaining the target interpolation speed of each track point, performing dynamic smoothing processing on the welding gun posture according to the target interpolation speed, correcting the target interpolation speed according to the predicted mechanical arm joint dynamic load, obtaining the corrected welding interpolation speed, dynamically adjusting the welding current and the wire feeding speed according to the corrected welding interpolation speed and the smoothed welding gun posture, and controlling the welding mechanical arm to perform welding work.

Description

Technical Field

[0001] This invention relates to the field of welding robot control technology, and more specifically, to a control method and a welding robot for welding scooter frames. Background Technology

[0002] With the rapid development of electric scooters, lightweight mobility scooters, and shared mobility devices, the production scale of scooter frames is constantly expanding. Scooter frames are usually formed by welding multiple sets of thin-walled metal tubes. Their structure generally has the characteristics of multiple spatial curved surfaces, complex tube intersections, continuous changes in welds, and high degree of lightweighting. Therefore, high requirements are placed on welding precision, weld consistency, and weld appearance.

[0003] Currently, automated welding operations are commonly performed using welding robots in the production of scooter frames. Existing welding robots typically consist of an industrial robot body, a welding power source, a welding torch, a positioner, and a robot controller. During the welding process, the robot controller drives the welding torch to move continuously along the weld seam of the frame according to a preset welding trajectory, and simultaneously controls the welding speed, welding posture, and welding parameters, thereby achieving automated welding.

[0004] In existing technologies, welding robots typically generate welding trajectory points through manual teaching or offline programming, and the robot controller then uses linear interpolation, circular interpolation, or spline interpolation to generate the welding torch motion trajectory. Simultaneously, the controller drives the robot to move along the weld seam according to a preset welding speed to complete the welding process.

[0005] However, most existing control methods employ fixed-speed interpolation or simple segmented speed control, lacking a dynamic speed coordination mechanism based on changes in weld curvature. When the robot moves to a region where weld curvature changes abruptly, problems such as instantaneous deceleration, motion jerking, and posture oscillation easily occur because the robot's movement direction and welding torch posture need to be adjusted rapidly. This leads to unstable welding heat input, easily causing uneven weld width, discontinuous fish-scale patterns, local burn-through, and welding deformation, which is particularly evident in the welding of thin-walled tubular components for scooter frames. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide a control method for welding robot for scooter frame and a welding robot, which solves the problems of motion jerking, posture oscillation and unstable welding heat input that are easy to occur in the area of ​​sudden change in weld curvature of the existing welding robot.

[0007] To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, this application provides a control method for a welding robot for a scooter frame. The method includes: identifying curvature abrupt change regions based on the curvature change rate of the weld trajectory, and performing dynamic feedforward speed planning to obtain the target interpolation speed for each trajectory point; dynamically smoothing the welding torch posture according to the target interpolation speed, and correcting the target interpolation speed according to the predicted dynamic load of the robot joint to obtain the corrected welding interpolation speed; dynamically adjusting the welding current and wire feeding speed according to the corrected welding interpolation speed and the smoothed welding torch posture, and controlling the welding robot to perform welding operations.

[0008] In one embodiment, identifying curvature abrupt change regions based on the rate of curvature change of the weld trajectory further includes: acquiring three-dimensional model data of the scooter frame, identifying the connection area between the pipes to be welded, and extracting the corresponding weld center trajectory curve; calculating the trajectory direction vector based on the weld center trajectory curve, and performing a cross product operation between the trajectory direction vector and the local surface normal direction of the pipe surface where the current trajectory point is located to generate a trajectory normal vector, thereby determining the welding torch running posture of each trajectory point; establishing a continuous welding trajectory model based on the weld center trajectory curve and the welding torch running posture, and discretizing the model at a preset sampling interval to generate a discrete trajectory point sequence containing spatial position, welding torch posture, and welding direction parameters.

[0009] In one embodiment, identifying curvature abrupt change regions based on the curvature change rate of the weld trajectory further includes: calculating a normalized first trajectory tangent vector and a second trajectory tangent vector based on the spatial coordinate differences between the current trajectory point and its previous and next trajectory points in the discrete trajectory point sequence; obtaining the change in the angle between the first trajectory tangent vector and the second trajectory tangent vector, and calculating the local curvature parameter corresponding to the current trajectory point based on the change in the angle; and calculating the curvature change rate parameter according to the ratio of the difference between adjacent local curvature parameters to the corresponding trajectory distance, based on the arrangement order of the trajectory points.

[0010] In one embodiment, identifying curvature abrupt change regions based on the curvature change rate of the weld trajectory includes: calculating the curvature fluctuation amplitude between adjacent trajectory points based on the curvature change rate parameter to generate a trajectory stability parameter; allocating a dynamic sampling window length to each trajectory point according to the trajectory stability parameter, and generating a curvature change continuity parameter based on the recalculated curvature change rate parameter; screening trajectory intervals with a curvature change continuity parameter greater than a preset threshold as candidate curvature abrupt change regions, and performing abrupt change level analysis based on the maximum curvature change rate and trajectory length of the candidate curvature abrupt change regions, and determining candidate regions with abrupt change level greater than a preset level threshold as weld curvature abrupt change regions.

[0011] In one embodiment, dynamic feedforward velocity planning is performed to obtain the target interpolation velocity for each trajectory point. This includes: constructing trajectory motion risk parameters by weighted summation based on the local curvature parameters, curvature change rate parameters, and curvature abrupt change region identifiers of the weld trajectory points; generating a basic interpolation velocity based on the trajectory motion risk parameters; obtaining the curvature change rate parameters and curvature abrupt change region distribution within the subsequent preset look-ahead range of the current trajectory point, generating feedforward deceleration parameters, and performing feedforward correction on the basic interpolation velocity to obtain the pre-planned interpolation velocity; and performing continuous smoothing processing based on the pre-planned interpolation velocity change gradient between adjacent trajectory points to obtain the target interpolation velocity for each trajectory point.

[0012] In one embodiment, the welding torch attitude is dynamically smoothed according to the target interpolation speed, including: calculating the attitude angle change and attitude change direction parameters between adjacent trajectory points based on the welding torch attitude angle parameters and the target interpolation speed of discrete trajectory points, and generating an initial attitude change rate sequence accordingly; counting the number of continuous trajectory points with the same attitude change direction within a preset window, generating an attitude change direction consistency parameter, and filtering out candidate abnormal attitude points with a consistency parameter lower than a preset threshold; correcting the initial attitude change rate of the candidate abnormal attitude points according to the local attitude change trend parameters within a preset range before and after the candidate abnormal attitude points to obtain a corrected attitude change rate, and generating a welding torch attitude interpolation trajectory based on the corrected attitude change rate.

[0013] In one embodiment, the target interpolation speed is corrected based on the predicted dynamic load of the manipulator joints to obtain the corrected welding interpolation speed. This includes: establishing a robot motion state sequence based on the target interpolation speed of each trajectory point and the welding torch attitude interpolation trajectory; performing inverse kinematics on the motion state sequence to obtain the joint angle, joint velocity, and joint acceleration corresponding to each trajectory point; calculating the inertial load component based on the joint inertia and joint acceleration, and combining it with the joint drive torque to obtain the dynamic load parameters of each joint.

[0014] In one embodiment, the target interpolation speed is corrected based on the predicted dynamic load of the robotic arm joints to obtain the corrected welding interpolation speed. The method further includes: normalizing and weighting the dynamic load parameters of each joint to generate joint load risk parameters; selecting trajectory points with joint load risk parameters greater than a preset threshold as high-load-risk trajectory points, and generating speed correction parameters based on their joint dynamic load parameters; and applying load compensation deceleration correction to the target interpolation speed according to the speed correction parameters to obtain the corrected welding interpolation speed.

[0015] In one embodiment, the welding current and wire feed speed are dynamically adjusted based on the corrected welding interpolation speed and the smoothed welding torch posture. This includes: calculating the unit length heat input parameters and welding torch posture disturbance parameters for each trajectory point based on the corrected welding interpolation speed and welding torch posture interpolation trajectory, and generating welding heat input change parameters; screening trajectory points with welding heat input change parameters greater than a preset threshold as abnormal heat input trajectory points, and generating welding current correction parameters and wire feed speed correction parameters proportionally based on the welding heat input change parameters; dynamically adjusting the current welding current and wire feed speed using the welding current correction parameters and wire feed speed correction parameters, and controlling the welding robot to perform welding operations according to the corrected welding interpolation speed, welding torch posture interpolation trajectory, and adjusted welding current and wire feed speed.

[0016] Secondly, this application provides a welding robot, comprising: The robotic arm body includes multiple joints and an end-effector welding torch; The sensor unit is used to acquire trajectory data of the weld seams of the scooter frame; The controller is connected to both the robotic arm body and the sensor unit, and is configured to execute the aforementioned robotic arm control method for welding scooter frames.

[0017] As can be seen from the above technical solutions, the embodiments of this application have the following advantages: Compared to traditional welding control methods that use fixed welding speeds and process parameters, this technical solution identifies areas of abrupt weld curvature changes based on the rate of curvature change of the weld trajectory. It then dynamically feeds forward speed planning based on these curvature changes, enabling proactive adjustment of the welding robot's operating speed. This reduces motion impact and speed abrupt changes in areas of sharp weld transitions. Simultaneously, by dynamically smoothing the welding torch posture based on the target interpolation speed and further correcting the interpolation speed using dynamic load predictions of the robot joints, it effectively reduces joint vibration, posture jitter, and trajectory tracking instability. Furthermore, by dynamically adjusting the welding current and wire feed speed based on the corrected welding interpolation speed and smoothed welding torch posture, it maintains a dynamic match between the welding heat input and the robot's motion. This effectively reduces heat input fluctuations, uneven weld formation, and welding spatter in areas of abrupt weld curvature changes, improving weld formation consistency, welding stability, and overall welding quality during the scooter frame welding process. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the control method for welding a robotic arm for a scooter frame provided in an embodiment of this application.

[0020] Figure 2 This is a schematic diagram of the curvature abrupt change region identification results based on a dynamic sampling window, provided in an embodiment of this application. Detailed Implementation

[0021] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] Reference Figure 1 The diagram shows a flowchart of the robotic arm control method for scooter frame welding provided by the present invention. The method includes the following steps: S1, acquire the trajectory data of the weld seam to be welded on the scooter frame, and generate a corresponding discrete trajectory point sequence based on the trajectory data, including: S101, acquire the three-dimensional model data of the scooter frame, and identify the connection area between the pipes to be welded based on the three-dimensional model data; Specifically, the spatial positional relationship of each tube in the scooter frame is obtained based on the 3D model data, and the corresponding tube connection area is identified based on the spatial intersection relationship between adjacent tubes.

[0023] S102, extract the corresponding weld boundary data based on the connection area, and generate the weld center trajectory curve based on the weld boundary data; S103, calculate the corresponding trajectory direction vector and trajectory normal vector based on the weld center trajectory curve, and determine the welding torch running posture according to the trajectory direction vector and trajectory normal vector; The step of calculating the corresponding trajectory direction vector based on the weld center trajectory curve includes: sequentially obtaining the spatial coordinate difference between adjacent trajectory sampling points according to the trajectory extension direction of the weld center trajectory curve, and performing unitization processing on the spatial coordinate change to obtain the trajectory direction vector of the corresponding trajectory sampling point.

[0024] The trajectory normal vector includes: obtaining the local surface normal direction of the pipe surface where the corresponding trajectory sampling point is located, and performing a cross product operation between the trajectory direction vector corresponding to the current trajectory sampling point and the local surface normal direction to generate a trajectory normal vector perpendicular to the trajectory direction vector.

[0025] The welding torch running direction is determined based on the trajectory direction vector, the welding torch tilting direction is determined based on the trajectory normal vector, and the welding torch attitude parameters of the corresponding trajectory points are generated based on the welding torch running direction and the welding torch tilting direction.

[0026] S104, Establish a corresponding continuous welding trajectory model based on the weld center trajectory curve and the welding torch running posture; S105, the continuous welding trajectory model is discretized according to the preset trajectory sampling interval to generate multiple continuously arranged trajectory sampling points; The discretization process includes: obtaining the total length of the trajectory corresponding to the continuous welding trajectory model; obtaining the number of trajectory samples based on the total length of the trajectory and the preset trajectory sampling interval; and extracting the corresponding trajectory sampling points sequentially along the continuous welding trajectory model according to the number of trajectory samples.

[0027] S106, Generate a corresponding discrete trajectory point sequence based on the spatial position parameters, welding torch attitude parameters and welding direction parameters corresponding to each of the trajectory sampling points.

[0028] S2, calculate the local curvature parameters and curvature change rate parameters corresponding to each trajectory point in the weld trajectory based on the discrete trajectory point sequence, and identify the weld curvature abrupt change region based on the curvature change rate parameters.

[0029] In this embodiment, the calculation of the local curvature parameters and curvature change rate parameters corresponding to each trajectory point in the weld trajectory based on the discrete trajectory point sequence includes: S201, according to the arrangement order of the discrete trajectory point sequence, obtain the spatial coordinate data between the current trajectory point and its previous and next trajectory points, and determine the spatial position change relationship between adjacent trajectory points based on the spatial coordinate data; S202, calculate the first trajectory tangent vector based on the spatial coordinate difference between the current trajectory point and its previous trajectory point, and calculate the second trajectory tangent vector based on the spatial coordinate difference between the current trajectory point and its next trajectory point; S203, the first trajectory tangent vector and the second trajectory tangent vector are respectively normalized to obtain the corresponding unit trajectory tangent vector; S204, obtain the change in the angle between the unit trajectory tangent vectors, and calculate the local curvature parameter corresponding to the current trajectory point based on the ratio of the change in the angle to the trajectory arc length between adjacent trajectory points; S205, according to the arrangement order of the discrete trajectory point sequence, obtain the difference between adjacent local curvature parameters, and calculate the corresponding curvature change rate parameter based on the ratio of the difference to the corresponding trajectory distance.

[0030] Further, identifying regions of abrupt changes in weld curvature based on the curvature change rate parameter includes: S206, Establish a corresponding trajectory stability analysis sequence based on the local curvature parameters and the curvature change rate parameters; S207, calculate the curvature fluctuation amplitude between adjacent trajectory points based on the trajectory stability analysis sequence, and generate the trajectory stability parameter of the corresponding trajectory point according to the reciprocal of the curvature fluctuation amplitude; The curvature fluctuation amplitude is the root mean square value of the difference between each curvature change rate parameter and the corresponding average curvature change rate parameter within the sampling window range. The smaller the curvature fluctuation amplitude, the higher the corresponding trajectory stability parameter.

[0031] S208, based on the trajectory stability parameters corresponding to each trajectory point, assign a corresponding dynamic sampling window length to each trajectory point; The process involves matching trajectory stability parameters with preset stability intervals; calling the corresponding window length parameter table based on the matching results, where different stability level intervals correspond to different numbers of trajectory sampling points; and assigning a corresponding dynamic sampling window length to each trajectory point based on the corresponding window length parameter table. The preset stability level intervals refer to multiple parameter intervals pre-divided according to the numerical range of the trajectory stability parameters, with different parameter intervals corresponding to different dynamic sampling window lengths.

[0032] S209, based on the dynamic sampling window length corresponding to each trajectory point, re-acquire multiple adjacent trajectory points within the corresponding window range, and recalculate the local curvature parameters and curvature change rate parameters of the corresponding trajectory points based on the multiple adjacent trajectory points; S2010, Generate corresponding curvature change continuity parameters based on the recalculated curvature change rate parameters, and determine the curvature change stability of the corresponding trajectory points based on the curvature change continuity parameters; The curvature change continuity parameter is the ratio of the number of trajectory points that continuously satisfy the curvature change rate growth condition to the corresponding trajectory length. The curvature change rate growth condition refers to the curvature change rate parameter corresponding to the current trajectory point continuously increasing relative to the curvature change rate parameter corresponding to the previous trajectory point.

[0033] S2011, filter the trajectory intervals where the curvature change continuity parameter is greater than the preset continuity threshold, and determine the corresponding trajectory intervals as candidate curvature change regions; The preset continuity threshold is determined in advance by statistically analyzing the mean, variance, or quantile interval of the continuity parameters corresponding to the continuous curvature change regions in the historical weld trajectory data.

[0034] S2012, based on the maximum curvature change rate parameter and the corresponding trajectory length parameter corresponding to the candidate curvature change region, a change level analysis is performed, and the candidate curvature change region with a change level greater than the preset level threshold is determined as the weld curvature change region.

[0035] The mutation level is the ratio of the maximum rate of change of curvature parameter to the corresponding trajectory length parameter.

[0036] like Figure 2 The figure shows a schematic diagram of the weld curvature abrupt change region identification results based on a dynamic sampling window in this invention. The horizontal axis represents the weld trajectory length, the vertical axis represents the curvature change rate parameter, and different shaded areas represent candidate curvature abrupt change regions after the dynamic sampling window adjustment. By comparing the fixed-window curvature analysis results with the dynamic-window analysis results of this invention, it can be seen that this invention can effectively suppress pseudo-abrupt peaks caused by local trajectory discrete noise, while also more continuously identifying real weld turning regions, thereby improving the accuracy and stability of weld curvature abrupt change identification.

[0037] It should be noted that, compared to conventional techniques that use fixed sampling windows and single curvature thresholds for weld abrupt change identification, this technique constructs a trajectory stability analysis sequence, dynamically generates trajectory stability parameters using curvature fluctuation amplitude, and adaptively adjusts the dynamic sampling window length based on these parameters. This allows for curvature analysis using trajectory sampling methods at different scales in different trajectory regions, effectively reducing the impact of discrete trajectory sampling noise on the curvature change rate calculation results. Simultaneously, by recalculating local curvature parameters and curvature change rate parameters, and combining these with curvature change continuity parameters to determine the trend of continuous curvature changes, it effectively distinguishes between pseudo-curvature abrupt changes caused by trajectory sampling errors and genuine weld curvature abrupt changes, avoiding misidentification problems caused by local noise spikes. Furthermore, by combining the maximum curvature change rate parameter and trajectory length parameter corresponding to candidate curvature abrupt change regions for abrupt change level analysis, the accuracy and stability of weld curvature abrupt change region identification results can be improved, thus providing more reliable trajectory basis data for subsequent dynamic speed planning, attitude adjustment, and heat input control of the welding robot.

[0038] S3, based on the curvature change rate parameter and the weld curvature abrupt change region, perform dynamic feedforward speed planning for the welding robot to obtain the target interpolation speed corresponding to each trajectory point, including: S301, obtain the local curvature parameters, curvature change rate parameters and corresponding weld curvature change area identification information corresponding to each trajectory point, and establish the corresponding trajectory dynamic feature sequence based on the local curvature parameters, curvature change rate parameters and weld curvature change area identification information. The weld curvature abrupt change region identification information is used to characterize whether the current trajectory point is located within the weld curvature abrupt change region.

[0039] S302, Calculate the trajectory motion risk parameters corresponding to each trajectory point based on the trajectory dynamic feature sequence, including: Obtain the local curvature parameters, curvature change rate parameters, and weld curvature abrupt change region identifier parameters corresponding to the current trajectory point; The local curvature parameter, the curvature change rate parameter, and the weld curvature abrupt change region identifier parameter are normalized respectively to obtain the corresponding normalized parameters; Based on the normalized local curvature parameters, curvature change rate parameters, and weld curvature abrupt change region identifier parameters, a weighted summation operation is performed to generate corresponding trajectory motion risk parameters. The weight coefficients in the weighted summation are predetermined by parameter calibration methods based on the statistical analysis results of the influence of local curvature parameters, curvature change rate parameters, and weld curvature abrupt change region identifier parameters on welding speed fluctuations in historical welding trajectory data.

[0040] S303 generates the corresponding basic interpolation velocity based on the trajectory motion risk parameters corresponding to each trajectory point, including: Obtain the preset reference welding speed corresponding to the welding robot; The corresponding velocity attenuation coefficient is obtained based on the trajectory motion risk parameters. The velocity attenuation coefficient is a velocity ratio adjustment parameter generated by linear mapping based on the trajectory motion risk parameters. The corresponding basic interpolation speed is generated based on the preset benchmark welding speed and the speed attenuation coefficient.

[0041] The basic interpolation speed refers to the initial trajectory running speed obtained after risk attenuation based on the trajectory motion risk parameters corresponding to the current trajectory point and a preset benchmark welding speed. The specific calculation formula is as follows:

[0042] In the formula, Based on interpolation speed, To preset the baseline welding speed, Let be the velocity attenuation coefficient, and 0 ≤ <1.

[0043] S304, obtain the curvature change rate parameters corresponding to multiple trajectory points within the subsequent preset look-ahead range of the current trajectory point, count the number and distribution of trajectory points corresponding to the weld curvature change abrupt region within the subsequent preset look-ahead range, and generate corresponding feedforward deceleration parameters based on the number, distribution, and curvature change rate growth trend of the trajectory points corresponding to the weld curvature change abrupt region. The process involves normalizing the number of trajectory points, corresponding trajectory distances, and curvature change rate growth trends corresponding to abrupt changes in weld curvature. Weighted summation is then performed on these normalized parameters to generate corresponding feedforward deceleration parameters. The weighting coefficients are determined by statistically analyzing the correlation coefficients between the number of trajectory points, corresponding trajectory distances, and curvature change rate growth trends in historical welding trajectory data and the welding speed fluctuation amplitude. Parameters with higher correlation coefficients have larger weighting coefficients. The weighting coefficients are determined based on the proportion of each correlation coefficient to the total correlation coefficient. The normalization process employs a maximum-minimum normalization method, mapping each parameter to the 0-1 interval according to its historical statistical range.

[0044] The preset look-ahead range is a preset trajectory length range located behind the current trajectory point along the direction of the discrete trajectory point sequence. It is calculated by multiplying the target interpolation speed by the speed response time and is used to ensure that the welding robot completes deceleration adjustment before reaching the region of abrupt change in weld curvature. The feedforward deceleration parameter refers to a predictive deceleration parameter generated based on the distribution of the region of abrupt change in weld curvature and the curvature change trend within the preset look-ahead range after the current trajectory point, used to adjust the operating speed of the welding robot in advance.

[0045] S305, Based on the aforementioned feedforward deceleration parameters, dynamically feedforward correct the basic interpolation velocity corresponding to the current trajectory point to obtain the corresponding pre-planned interpolation velocity, including: Obtain the basic interpolation velocity and corresponding feedforward deceleration parameters corresponding to the current trajectory point; Calculate the corresponding speed correction based on the feedforward deceleration parameters; The specific formula for calculating the speed correction is as follows:

[0046] In the formula, For feedforward deceleration parameters, This is the speed correction amount.

[0047] Based on the speed correction amount, the basic interpolation speed corresponding to the current trajectory point is decelerated and corrected in advance to obtain the corresponding pre-planned interpolation speed. The pre-planned interpolation speed refers to the trajectory pre-adjustment speed obtained based on the basic interpolation speed and after pre-deceleration correction using feedforward deceleration parameters. The specific calculation formula is as follows:

[0048] In the formula, To pre-plan the interpolation speed, This is the speed correction amount.

[0049] S306, generating corresponding velocity continuity constraint parameters based on the pre-planned interpolation velocity changes between adjacent trajectory points, and performing continuous smoothing processing on the pre-planned interpolation velocity according to the velocity continuity constraint parameters to obtain the corresponding target interpolation velocity, including: Obtain the pre-planned interpolation speed difference between adjacent trajectory points; Calculate the velocity gradient between adjacent trajectory points; The velocity change gradient is the ratio of the pre-planned interpolation velocity difference between adjacent trajectory points to the corresponding trajectory distance.

[0050] The velocity change gradient is compared with a preset velocity change gradient threshold. The preset speed change gradient threshold is determined in advance by threshold calibration based on the statistical results of speed change gradients corresponding to adjacent trajectory points in historical welding trajectory data, combined with the maximum allowable acceleration and maximum impact parameters of the welding robot.

[0051] When the velocity change gradient is greater than the preset velocity change gradient threshold, the pre-planned interpolation velocity of the corresponding trajectory point is gradually transitioned to limit the velocity change amplitude between adjacent trajectory points and obtain the corresponding target interpolation velocity.

[0052] Specifically, when the smoothed speed is greater than the preset safe speed corresponding to the current trajectory point, the preset safe speed is determined as the target interpolation speed for the corresponding trajectory point; when the smoothed speed is less than or equal to the preset safe speed, the smoothed interpolation speed is determined as the target interpolation speed for the corresponding trajectory point. The preset safe speed is an allowable operating speed predetermined based on the local curvature parameters, curvature change rate parameters, and weld curvature abrupt change regions corresponding to the current trajectory point. By adding a safe speed constraint after speed continuity smoothing, the speed smoothing process can be prevented from weakening the feedforward deceleration effect, ensuring that the welding robot completes the expected deceleration before entering the weld curvature abrupt change region, thus balancing speed continuity and trajectory tracking safety.

[0053] The target interpolation speed refers to the final operating speed obtained after constraining and smoothing the pre-planned interpolation speed, which is used to control the actual trajectory interpolation motion of the welding robot.

[0054] It should be noted that, compared with conventional welding control schemes that employ fixed speed interpolation and fixed deceleration strategies, this technical solution constructs trajectory motion risk parameters by introducing local curvature parameters, curvature change rate parameters, and weld curvature abrupt change region identifier parameters. Combined with the distribution of weld curvature abrupt change regions within the look-ahead range, dynamic feedforward deceleration control is performed. This allows the welding robot to complete speed adjustments before entering the weld curvature abrupt change region, effectively reducing motion jerking and attitude oscillations caused by instantaneous rapid deceleration. Simultaneously, through continuous smoothing processing based on velocity change gradients, the amplitude of velocity abrupt changes between adjacent trajectory points can be limited, improving the trajectory continuity and motion stability of the welding robot during operation, thereby enhancing the stability of welding heat input and the weld formation quality of the scooter frame.

[0055] S4, calculate the welding torch attitude change rate based on the target interpolation speed, and perform dynamic smoothing processing on the welding torch attitude transition process according to the welding torch attitude change rate to obtain the corresponding welding torch attitude interpolation trajectory, including: S401, obtain the welding gun attitude angle parameters and target interpolation speed corresponding to each trajectory point in the discrete trajectory point sequence, and establish the corresponding attitude change analysis sequence based on the welding gun attitude angle parameters corresponding to each trajectory point; The welding torch attitude angle parameters include the pitch angle, yaw angle, and roll angle parameters corresponding to the welding torch.

[0056] S402, based on the attitude change analysis sequence, calculate the attitude angle change and attitude change direction parameters between adjacent trajectory points, including: Obtain the pitch angle change, yaw angle change, and roll angle change between adjacent trajectory points; The corresponding comprehensive attitude angle change is generated by summing the squares of the pitch angle change, yaw angle change, and roll angle change and then taking the square root. Generate the corresponding attitude change direction parameters based on the changing trend of the comprehensive attitude angle change between adjacent trajectory points; Specifically, the difference between the comprehensive attitude angle change corresponding to the current trajectory point and the comprehensive attitude angle change corresponding to the previous trajectory point is obtained, and the difference is compared with a preset attitude direction determination dead zone threshold. The preset attitude direction determination dead zone threshold is determined in advance based on the statistical results of the normal attitude fluctuation range in historical welding trajectory data and in combination with the attitude sampling accuracy.

[0057] If the difference is greater than the preset attitude direction determination dead zone threshold, the attitude change direction parameter is determined to be an increasing direction. If the difference is less than the opposite of the preset attitude direction determination dead zone threshold, the attitude change direction parameter is determined to be a decreasing direction. When the difference is within the range corresponding to the preset attitude direction determination dead zone threshold, the attitude change direction parameter is determined to be a maintaining direction. The maintaining direction indicates that the current comprehensive attitude angle change is within the normal fluctuation range and does not participate in the attitude abnormal change determination.

[0058] S403, calculates the initial attitude change rate sequence for each trajectory point based on the attitude angle change between adjacent trajectory points and the corresponding target interpolation velocity, including: Obtain the trajectory distance parameters between adjacent trajectory points and the target interpolation speed; Calculate the corresponding trajectory running time parameters based on the trajectory distance parameters and the target interpolation velocity; The initial attitude change rate is generated based on the ratio of the attitude angle change to the corresponding trajectory running time parameter.

[0059] S404, based on the initial attitude change rate sequence, statistically analyze the consistency parameters of the attitude change direction corresponding to multiple consecutive trajectory points, including: The target interpolation speed corresponding to the current trajectory point and the preset attitude analysis time length are obtained. The preset attitude analysis time length is the time parameter corresponding to the welding robot attitude control system completing one attitude adjustment response. The dynamic attitude analysis window length is calculated based on the product of the target interpolation velocity and the preset attitude analysis time length. The number of corresponding window trajectory points is determined based on the dynamic attitude analysis window length and trajectory sampling interval. Obtain the attitude change direction parameters of multiple adjacent trajectory points within the range of the number of trajectory points in the window before and after the current trajectory point; Count the number of consecutive trajectory points with the same direction of attitude change; The attitude change direction consistency parameter is generated based on the ratio of the number of continuous trajectory points with the same attitude change direction to the total number of trajectory points within the window.

[0060] It should be noted that, to ensure the consistency of attitude continuity analysis results under different operating speeds, a method of dynamically adjusting the attitude analysis window length based on the target interpolation speed is adopted, so that the analysis time range corresponding to the window remains basically consistent. Compared with the window method with a fixed number of trajectory points or a fixed trajectory distance, this embodiment can avoid the attitude continuity evaluation deviation caused by the change of analysis window scale under high-speed and low-speed operation of the welding robot, and improve the accuracy and stability of the candidate abnormal attitude point identification results. S405, compare the attitude change direction consistency parameter corresponding to each trajectory point with the preset continuity judgment threshold, and select trajectory points whose attitude change direction consistency parameter is lower than the preset continuity judgment threshold as candidate abnormal attitude points. The preset continuity judgment threshold is determined in advance based on the statistical results of the consistency parameters of the posture change direction corresponding to the normal welding posture change process in historical welding trajectory data, through the statistical mean and variance range.

[0061] S406, obtain the initial attitude change rate corresponding to multiple adjacent trajectory points within a preset trajectory range before and after the candidate abnormal attitude point, and calculate the corresponding local attitude change trend parameters based on the initial attitude change rate corresponding to the multiple adjacent trajectory points, including: The local attitude change trend parameters are generated by calculating the moving average based on the initial attitude change rate corresponding to multiple adjacent trajectory points. The local attitude change trend parameter is the average of the initial attitude change rates of multiple adjacent trajectory points.

[0062] S407, Based on the local attitude change trend parameters, perform trend consistency correction on the initial attitude change rate corresponding to the candidate abnormal attitude points to obtain the corresponding corrected attitude change rate, including: Calculate the difference between the initial attitude change rate and the local attitude change trend parameter corresponding to the candidate abnormal attitude point; The difference is compared with a preset attitude deviation threshold. When the difference is greater than the preset attitude deviation threshold, the initial attitude change rate corresponding to the candidate abnormal attitude point is proportionally corrected based on the local attitude change trend parameter in order to reduce the amplitude of local abnormal attitude fluctuation. The preset attitude deviation threshold is determined based on the statistical results of the difference between the initial attitude change rate and the local attitude change trend parameter under normal welding conditions in historical welding trajectory data, and in combination with the maximum attitude fluctuation range allowed by the welding robot.

[0063] The specific calculation formula for the corrected attitude change rate is as follows:

[0064] In the formula, To correct the rate of attitude change, The initial attitude change rate, These are parameters representing the local attitude change trend. To predetermine the trend correction coefficient, statistical analysis was performed on the initial attitude change rate and corresponding local attitude change trend parameters in historical welding trajectory data. Different candidate β values ​​were iterated through, and the minimum attitude fluctuation amplitude was used as the optimization objective to determine β. .

[0065] S408, based on the corrected attitude change rate corresponding to each trajectory point, regenerate the corresponding welding gun attitude smoothing adjustment parameters, and perform dynamic smoothing processing on the welding gun attitude transition process based on the welding gun attitude smoothing adjustment parameters to generate the corresponding welding gun attitude interpolation trajectory.

[0066] The welding torch attitude interpolation trajectory refers to the continuous attitude control trajectory generated after continuous interpolation calculation based on the welding torch attitude parameters corresponding to each discrete trajectory point. It is used to control the welding torch to perform continuous and smooth attitude adjustment along the weld seam trajectory.

[0067] Based on the welding torch posture interpolation trajectory, the welding robot is controlled to perform weld seam tracking welding motion, so as to reduce local oscillations and sudden changes in the welding torch posture, and improve the continuity of the welding torch posture and the welding stability.

[0068] It should be noted that, compared with conventional attitude interpolation control schemes that directly calculate the attitude change rate based on adjacent trajectory points, this technical scheme introduces an attitude change direction consistency parameter to perform continuity analysis on the attitude change trend in continuous trajectory segments, and combines local attitude change trend parameters to perform trend consistency correction on candidate abnormal attitude points. This can effectively suppress the amplification of abnormal attitude change rate caused by local sampling errors, trajectory discrete fluctuations, or attitude abrupt changes, thereby reducing local oscillations and attitude abrupt changes in the welding torch attitude. At the same time, by dynamically smoothing the welding torch attitude transition process, the continuity and stability of the welding torch attitude interpolation trajectory can be improved, thereby enhancing the trajectory tracking accuracy and welding quality stability of the welding robot in the region of abrupt change in weld curvature.

[0069] S5. Based on the target interpolation speed and the welding torch attitude interpolation trajectory, predict the dynamic load parameters of each joint of the welding robot, and correct the target interpolation speed according to the dynamic load parameters to obtain the corrected welding interpolation speed.

[0070] In this embodiment, the dynamic load parameters of each joint of the welding robot are predicted based on the target interpolation speed and the welding torch attitude interpolation trajectory, including: S501, obtain the target interpolation velocity and welding gun attitude interpolation trajectory corresponding to each trajectory point, and establish the corresponding robot motion state sequence based on the target interpolation velocity and welding gun attitude interpolation trajectory; The robot motion state sequence includes the welding torch spatial position parameters, welding torch attitude parameters, and target interpolation velocity parameters corresponding to each trajectory point.

[0071] S502, Based on the robot motion state sequence, perform joint kinematics solution on the welding manipulator to obtain the joint motion parameter sequence corresponding to each trajectory point, including: Obtain the spatial position parameters and attitude parameters of the welding torch corresponding to each trajectory point; Based on the kinematic model of the welding robot, the spatial position parameters and attitude parameters of the welding torch corresponding to each trajectory point are solved by inverse kinematics to obtain the corresponding joint angle parameters. Obtain the change in joint angle between adjacent trajectory points; Obtain the maximum permissible joint velocity parameters for each joint; The theoretical joint running time required for each joint to complete the angle change is calculated based on the ratio of the absolute value of the angle change of each joint to the corresponding maximum allowable joint velocity parameter. The maximum value among the theoretical joint running times of each joint is selected as the joint synchronization running time parameter corresponding to the current trajectory segment; The joint velocity parameters corresponding to each joint are calculated based on the ratio of the change in joint angle to the synchronous running time parameter of the joint. Obtain the joint velocity changes corresponding to adjacent trajectory segments; The joint acceleration parameters corresponding to each joint are calculated based on the ratio of the change in joint velocity to the joint synchronous running time parameter. The joint angle parameters include joint angle parameters, joint velocity parameters, and joint acceleration parameters.

[0072] S503 predicts the corresponding joint dynamic load parameters based on the joint angle parameters, joint velocity parameters, and joint acceleration parameters of each joint, including: Obtain the joint inertia parameters, robotic arm load mass parameters, and link structure parameters for each joint; The inertial load component is calculated based on the product of the joint inertia parameter and the joint acceleration parameter. The corresponding gravity load component is calculated based on the robot arm's load mass parameters, joint angle parameters, and gravitational acceleration parameters.

[0073] In the formula, This represents the gravity load component corresponding to the i-th joint. Let be the equivalent load mass corresponding to the rear end mechanism of the i-th joint. It is the acceleration due to gravity. Let be the equivalent force arm length from the load center of gravity to the rotation axis of the i-th joint. Let be the joint angle parameter corresponding to the i-th joint.

[0074] Calculate the corresponding motion coupling load components based on joint velocity parameters and the kinematic coupling relationship of the robotic arm;

[0075] In the formula, This represents the motion coupling load component corresponding to the i-th joint. Let be the motion coupling coefficient corresponding to the i-th joint, provided by the robot manufacturer or obtained through robot dynamics parameters. Let be the joint velocity parameter corresponding to the i-th joint.

[0076] The corresponding joint dynamic load parameters are generated by summing the inertial load component, the gravity load component, and the motion coupling load component.

[0077] The joint dynamic load parameter is used to characterize the dynamic load degree of each joint of the welding robot under the current motion state.

[0078] Further, the target interpolation speed is corrected based on the dynamic load parameters to obtain the corrected welding interpolation speed, including: S504 generates corresponding joint load risk parameters based on the dynamic load parameters of each joint, including: The maximum and minimum value normalization method is used to normalize the dynamic load parameters corresponding to each joint, so as to map the corresponding parameters to the 0 to 1 range; The load weight coefficients for each joint are obtained. These load weight coefficients are determined by statistically analyzing the correlation coefficients between the dynamic load parameters of each joint and the vibration amplitude of the robot arm in historical welding operation data. The corresponding joint load risk parameters are generated by performing a weighted summation operation based on the normalized dynamic load parameters and the corresponding load weight coefficients.

[0079] S505, compare the joint load risk parameter with the preset joint load threshold, and select trajectory points whose joint load risk parameter is greater than the preset joint load threshold as high load risk trajectory points. The preset joint load threshold is determined based on the statistical results of the dynamic load parameters of each joint during historical welding operations, combined with the maximum allowable joint load range and maximum servo drive capability of the welding robot.

[0080] S506, based on the joint dynamic load parameters corresponding to high load risk trajectory points, converts the load magnitude into a speed correction parameter in the range of 0 to 1 through a linear mapping method, so as to characterize the relationship that the higher the joint load, the greater the corresponding speed correction magnitude. S507, Based on the speed correction parameters, the target interpolation speed of the corresponding trajectory point is corrected by load compensation to obtain the corrected welding interpolation speed, including: Calculate the corresponding speed compensation amount based on the speed correction parameters; The specific formula for calculating the speed compensation amount is as follows:

[0081] In the formula, For speed compensation amount, For speed correction parameters, The target interpolation speed.

[0082] Based on the speed compensation amount, the target interpolation speed is decelerated and corrected to obtain the corrected welding interpolation speed; The specific calculation formula for the corrected welding interpolation speed is as follows:

[0083] In the formula, This is the corrected welding interpolation speed.

[0084] The speed correction parameter takes effect only once in the current trajectory sampling period and re-participates in the joint kinematics calculation and joint dynamic load prediction in the next trajectory sampling period to form a discretely updated closed-loop speed control process.

[0085] It should be noted that, compared to traditional welding robot control schemes that rely solely on trajectory geometry parameters for speed interpolation control, this technical solution further combines the welding torch attitude interpolation trajectory with the dynamic motion state of each joint of the welding robot. It predicts and analyzes the dynamic load on each joint during welding operation and performs real-time compensation and correction of the target interpolation speed based on dynamic load parameters. This allows for early reduction of the operating speed under high dynamic load conditions in areas of abrupt weld curvature changes, effectively mitigating issues such as sudden acceleration changes in robot joints, joint vibration, and servo shock. Simultaneously, by establishing joint load risk parameters and dynamically correcting high-load-risk trajectory points, it improves the motion stability, trajectory tracking accuracy, and weld formation consistency of the welding robot under complex weld trajectories, thereby enhancing overall welding quality and equipment operational reliability.

[0086] S6. Calculate the welding heat input variation parameters based on the corrected welding interpolation speed and welding torch posture interpolation trajectory, and dynamically adjust the welding current parameters and wire feed speed parameters according to the welding heat input variation parameters.

[0087] In this embodiment, welding heat input variation parameters are calculated based on the corrected welding interpolation speed and welding torch attitude interpolation trajectory, and welding current parameters and wire feed speed parameters are dynamically adjusted according to the welding heat input variation parameters, including: S601, obtain the corrected welding interpolation speed and welding torch posture interpolation trajectory corresponding to each trajectory point, and establish the corresponding welding heat input analysis sequence based on the corrected welding interpolation speed and welding torch posture interpolation trajectory. The welding heat input analysis sequence includes welding interpolation speed parameters, welding torch attitude angle parameters, and welding torch attitude change rate corresponding to each trajectory point.

[0088] S602, Calculate the welding heat input parameters corresponding to each trajectory point based on the welding heat input analysis sequence, including: Obtain the welding voltage parameters, welding current parameters, and corrected welding interpolation speed corresponding to the current trajectory point; The corresponding heat input parameters per unit length are calculated based on welding voltage parameters, welding current parameters, and the corrected welding interpolation speed. The heat input parameter is used to characterize the heat input variation trend corresponding to a unit weld length, and the specific calculation formula is as follows:

[0089] In the formula, For thermal input parameters, The welding thermal efficiency coefficient can be a preset constant within the range of industry standards or experience, selected according to the welding process type (such as MIG / MAG, TIG, etc.). These are the welding voltage parameters. These are welding current parameters. This is the welding torch attitude correction factor, used to characterize the welding torch pitch angle. Yaw angle and roll angle It affects the direction of the electric arc and the heated area of ​​the molten pool, and is used to correct the spatial directionality of heat input per unit length.

[0090] The welding torch attitude correction factor is calculated by normalizing the pitch angle, yaw angle and roll angle of the welding torch, and then weighting and fusing them based on the cosine value of the spatial angle between them and the preset reference attitude. The weight coefficient of each attitude component is determined statistically based on the influence of the corresponding attitude change on the weld penetration deviation in historical welding data, and the value range of the welding torch attitude correction factor is guaranteed to be within the preset range through normalization constraints.

[0091] S603, calculate the corresponding welding gun attitude disturbance parameters based on the welding gun attitude interpolation trajectory; Specifically, the historical minimum and maximum values ​​of the welding torch attitude change rate are obtained, and the maximum and minimum values ​​are linearly normalized to map the current welding torch attitude change rate to the 0-1 interval according to the corresponding historical statistical range, thereby obtaining the corresponding welding torch attitude disturbance parameters.

[0092] S604 generates corresponding welding heat input variation parameters based on unit length heat input parameters and welding torch posture disturbance parameters; The maximum and minimum value normalization method is used to normalize the thermal input parameters per unit length and the welding torch attitude disturbance parameters respectively, so as to map the corresponding parameters to the 0 to 1 range; The heat input weighting coefficient and the attitude disturbance weighting coefficient are obtained, wherein the heat input weighting coefficient and the attitude disturbance weighting coefficient are determined by statistically analyzing the correlation coefficients between the heat input parameters per unit length and the welding torch attitude disturbance parameters and the weld penetration fluctuation amplitude in historical welding process data. The welding heat input variation parameters are generated by performing a weighted summation operation based on the normalized unit length heat input parameters and the welding torch attitude disturbance parameters. These welding heat input variation parameters are used as control parameters to characterize the degree of heat input fluctuation during the welding process.

[0093] S605, compare the welding heat input change parameter with the preset heat input fluctuation threshold, and select trajectory points whose welding heat input change parameter is greater than the preset heat input fluctuation threshold as heat input abnormal trajectory points; The preset heat input fluctuation threshold is determined based on the statistical results of welding heat input variation parameters under normal welding conditions in historical welding process data, combined with the pre-determined allowable heat input fluctuation range of the weld.

[0094] S606 generates corresponding welding current correction parameters and wire feed speed correction parameters based on the welding heat input change parameters corresponding to the abnormal heat input trajectory points, including: Obtain the welding heat input change parameters corresponding to the abnormal heat input trajectory points; The corresponding welding current correction parameter is generated based on the product of the welding heat input variation parameter and the preset current correction ratio coefficient. The corresponding wire feed speed correction parameter is generated by multiplying the welding heat input variation parameter with the preset wire feed correction ratio coefficient. The method for determining the preset current correction ratio coefficient and the preset wire feed speed correction ratio coefficient includes: establishing a first linear regression model between the welding heat input variation parameter and the weld penetration fluctuation amplitude, and a second linear regression model between the welding heat input variation parameter and the weld reinforcement fluctuation amplitude, and determining the corresponding correction ratio coefficient based on the fitting slope of each regression model.

[0095] S607, Based on the welding current correction parameters, the current welding current parameters are dynamically corrected to obtain the corresponding target welding current parameters; The specific calculation formula for the target welding current parameter is as follows:

[0096] In the formula, For the target welding current parameters, For welding current correction parameters, This refers to the current welding current parameter.

[0097] S608, Based on the wire feeding speed correction parameter, the current wire feeding speed parameter is dynamically corrected to obtain the corresponding target wire feeding speed parameter; The specific calculation formula for the target wire feeding speed parameter is as follows:

[0098] In the formula, The target wire feed speed parameter is the target operating parameter used to control the wire feed speed. For wire feed speed correction parameters, This refers to the current wire feeding speed parameter.

[0099] The welding current correction parameter and wire feed speed correction parameter are signed proportional adjustment values. Their positive and negative values ​​correspond to decreasing adjustment when the welding heat input is too high and increasing adjustment when the welding heat input is too low, respectively, thereby realizing bidirectional closed-loop compensation control based on heat input error.

[0100] It should be noted that by combining the corrected welding interpolation speed and the welding torch posture interpolation trajectory, a comprehensive analysis of the changes in heat input per unit length and the disturbance of the welding torch posture during the welding process is performed, and corresponding welding heat input change parameters are dynamically generated. This can more accurately reflect the heat input fluctuation state during the welding process. At the same time, by dynamically proportionally correcting the welding current parameters and wire feed speed parameters based on the welding heat input change parameters, the problem of local heat input instability caused by changes in welding speed, welding torch posture fluctuations, and sudden changes in weld curvature can be effectively reduced. This improves the uniformity of weld penetration, the uniformity of weld formation, and the stability of the welding process, and reduces the probability of welding defects such as welding spatter, burn-through, and incomplete welds.

[0101] S7 controls the welding robot to perform scooter frame welding operations according to the corrected welding interpolation speed, welding torch posture interpolation trajectory, dynamically adjusted welding current parameters, and wire feed speed parameters, including: The system acquires the corrected welding interpolation speed, welding torch posture interpolation trajectory, dynamically adjusted welding current parameters, and wire feed speed parameters. Based on the corrected welding interpolation speed and welding torch posture interpolation trajectory, it generates corresponding end-effector motion control commands for the welding robot. Based on the welding torch posture interpolation trajectory, it performs real-time interpolation calculations on the target motion position, target motion speed, and target motion acceleration corresponding to each joint of the welding robot, and controls the welding robot to drive each joint to perform welding torch trajectory tracking motion according to the corresponding target motion parameters. Simultaneously, based on the dynamically adjusted welding current parameters, it controls the welding power supply to adjust the corresponding welding output current in real time, and based on the dynamically adjusted wire feed speed parameters, it controls the wire feeding mechanism to adjust the wire feed speed in real time. During the continuous movement of the welding robot along the weld seam trajectory, it synchronously executes welding torch spatial position control, welding torch posture control, welding current control, and wire feed speed control in real time. This ensures that the welding torch stably completes the scooter frame weld seam welding operation according to the corrected welding interpolation speed and welding torch posture interpolation trajectory, thereby reducing motion oscillations, heat input fluctuations, and uneven weld seam formation in the weld seam curvature abrupt change area, improving welding process stability and weld seam formation quality.

[0102] This embodiment provides a welding robot, including: a robot body, a sensor unit, and a controller.

[0103] The robotic arm is a multi-jointed, serially connected industrial robot with a welding torch mounted at its end. Each joint of the robotic arm is driven by a corresponding servo motor and equipped with an encoder for real-time feedback of joint angles and speeds.

[0104] The sensor unit includes a vision sensor or a laser structured light sensor, which is installed on one side of the welding torch to acquire trajectory data of the weld seam of the scooter frame in real time.

[0105] The controller is an industrial computer or an embedded motion controller, electrically connected to the servo driver of the robotic arm body and the sensor unit, respectively. The controller is configured to execute the aforementioned control method for the robotic arm used for welding scooter frames.

[0106] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0107] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0108] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0109] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for controlling a robotic arm for welding scooter frames, characterized in that, include: Based on the curvature change rate of the weld trajectory, regions of curvature abrupt change are identified, and dynamic feedforward velocity planning is performed to obtain the target interpolation velocity for each trajectory point. The welding torch posture is dynamically smoothed according to the target interpolation speed, and the target interpolation speed is corrected according to the predicted dynamic load of the robot joint to obtain the corrected welding interpolation speed. Based on the corrected welding interpolation speed and the smoothed welding torch posture, the welding current and wire feed speed are dynamically adjusted, and the welding robot is controlled to perform the welding operation.

2. The method according to claim 1, characterized in that, The method for identifying curvature abrupt change regions based on the curvature change rate of the weld trajectory also includes: Obtain the 3D model data of the scooter frame, identify the connection area between the pipes to be welded, and extract the corresponding weld center trajectory curve; The trajectory direction vector is calculated based on the weld center trajectory curve, and the trajectory direction vector is cross-multiplied with the local surface normal direction of the pipe surface where the current trajectory point is located to generate the trajectory normal vector, thereby determining the welding gun running posture of each trajectory point. A continuous welding trajectory model is established based on the weld center trajectory curve and the welding torch running posture, and a discrete trajectory point sequence containing spatial position, welding torch posture and welding direction parameters is generated by discretization according to the preset sampling interval.

3. The method according to claim 2, characterized in that, The method for identifying curvature abrupt change regions based on the curvature change rate of the weld trajectory also includes: Based on the spatial coordinate differences between the current trajectory point and its previous and next trajectory points in the discrete trajectory point sequence, the normalized first trajectory tangent vector and second trajectory tangent vector are calculated respectively. Obtain the change in the angle between the first trajectory tangent vector and the second trajectory tangent vector, and calculate the local curvature parameter corresponding to the current trajectory point based on the change in the angle; The curvature change rate parameter is calculated based on the ratio of the difference between adjacent local curvature parameters to the corresponding trajectory distance, according to the arrangement order of the trajectory points.

4. The method according to claim 3, characterized in that, The method for identifying curvature abrupt change regions based on the rate of curvature change of the weld trajectory includes: The curvature fluctuation amplitude between adjacent trajectory points is calculated based on the curvature change rate parameter to generate trajectory stability parameters. The trajectory stability parameter is used to assign a dynamic sampling window length to each trajectory point, and a curvature change continuity parameter is generated based on the recalculated curvature change rate parameter. Trajectory intervals with curvature change continuity parameters greater than a preset threshold are selected as candidate curvature change regions. Change level analysis is performed based on the maximum curvature change rate and trajectory length of the candidate curvature change regions, and candidate regions with change levels greater than a preset level threshold are determined as weld curvature change regions.

5. The method according to claim 1, characterized in that, The process of performing dynamic feedforward velocity planning to obtain the target interpolation velocity for each trajectory point includes: Based on the local curvature parameters, curvature change rate parameters, and curvature abrupt change region identifiers of the weld trajectory points, trajectory motion risk parameters are constructed by weighted summation, and a basic interpolation velocity is generated based on the trajectory motion risk parameters. Obtain the curvature change rate parameter and curvature abrupt change region distribution within the preset look-ahead range of the current trajectory point, generate feedforward deceleration parameters, and perform feedforward correction on the basic interpolation velocity to obtain the pre-planned interpolation velocity; The target interpolation velocity for each trajectory point is obtained by performing continuous smoothing based on the pre-planned interpolation velocity change gradient between adjacent trajectory points.

6. The method according to claim 1, characterized in that, The dynamic smoothing of the welding torch posture based on the target interpolation speed includes: Based on the welding gun attitude angle parameters and target interpolation velocity of discrete trajectory points, the attitude angle change and attitude change direction parameters between adjacent trajectory points are calculated, and an initial attitude change rate sequence is generated accordingly. The number of continuous trajectory points with the same attitude change direction within a preset window is counted, an attitude change direction consistency parameter is generated, and candidate abnormal attitude points with a consistency parameter lower than a preset threshold are selected. Based on the local attitude change trend parameters within a preset range before and after the candidate abnormal attitude point, the initial attitude change rate of the candidate abnormal attitude point is corrected for trend consistency to obtain the corrected attitude change rate, and the welding gun attitude interpolation trajectory is generated based on the corrected attitude change rate.

7. The method according to claim 1, characterized in that, The step of correcting the target interpolation speed based on the predicted dynamic load of the robotic arm joint to obtain the corrected welding interpolation speed includes: Based on the target interpolation velocity and welding gun attitude interpolation trajectory of each trajectory point, a sequence of robot motion states is established. The inverse kinematics solution is performed on the motion state sequence to obtain the joint angles, joint velocities and joint accelerations corresponding to each trajectory point; The inertial load component is calculated based on the joint inertia and joint acceleration, and combined with the joint driving torque, the dynamic load parameters of each joint are obtained.

8. The method according to claim 7, characterized in that, The step of correcting the target interpolation speed based on the predicted dynamic load of the robotic arm joint to obtain the corrected welding interpolation speed further includes: The dynamic load parameters of each joint are normalized and weighted and summed to generate joint load risk parameters; Trajectory points with joint load risk parameters greater than a preset threshold are selected as high load risk trajectory points, and speed correction parameters are generated based on their joint dynamic load parameters. The target interpolation speed is corrected by load compensation and deceleration based on the speed correction parameters to obtain the corrected welding interpolation speed.

9. The method according to claim 8, characterized in that, The step of dynamically adjusting the welding current and wire feed speed based on the corrected welding interpolation speed and the smoothed welding torch posture includes: Based on the corrected welding interpolation speed and welding torch attitude interpolation trajectory, the unit length thermal input parameters and welding torch attitude disturbance parameters of each trajectory point are calculated, and welding thermal input variation parameters are generated. Trajectory points where the welding heat input change parameters are greater than a preset threshold are selected as abnormal heat input trajectory points, and welding current correction parameters and wire feed speed correction parameters are generated proportionally based on the welding heat input change parameters. The welding current and wire feed speed are dynamically adjusted using the welding current correction parameters and wire feed speed correction parameters, and the welding robot is controlled to perform welding operations according to the corrected welding interpolation speed, welding torch posture interpolation trajectory, and adjusted welding current and wire feed speed.

10. A welding robot, characterized in that, include: The robotic arm body includes multiple joints and an end-effector welding torch; The sensor unit is used to acquire trajectory data of the weld seams of the scooter frame; The controller is connected to both the robotic arm body and the sensor unit, and is configured to perform the robotic arm control method for scooter frame welding as described in any one of claims 1 to 9.

Images